IoU-Adaptive Deformable R-CNN: Make Full Use of IoU for Multi-Class Object Detection in Remote Sensing Imagery

被引:113
作者
Yan, Jiangqiao [1 ,2 ,3 ]
Wang, Hongqi [1 ,3 ]
Yan, Menglong [1 ,3 ]
Diao, Wenhui [1 ,3 ]
Sun, Xian [1 ,3 ]
Li, Hao [1 ,3 ]
机构
[1] Chinese Acad Sci, Inst Elect, Beijing 100190, Peoples R China
[2] Univ Chinese Acad Sci, Sch Elect Elect & Commun Engn, Beijing 100190, Peoples R China
[3] Chinese Acad Sci, Inst Elect, Key Lab Technol Geospatial Informat Proc & Applic, Beijing 100190, Peoples R China
基金
中国国家自然科学基金;
关键词
remote sensing imagery; anchor matching; Cascade R-CNN; IoU-based weighted loss; non-maximum suppression; VEHICLE DETECTION;
D O I
10.3390/rs11030286
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Recently, methods based on Faster region-based convolutional neural network (R-CNN) have been popular in multi-class object detection in remote sensing images due to their outstanding detection performance. The methods generally propose candidate region of interests (ROIs) through a region propose network (RPN), and the regions with high enough intersection-over-union (IoU) values against ground truth are treated as positive samples for training. In this paper, we find that the detection result of such methods is sensitive to the adaption of different IoU thresholds. Specially, detection performance of small objects is poor when choosing a normal higher threshold, while a lower threshold will result in poor location accuracy caused by a large quantity of false positives. To address the above issues, we propose a novel IoU-Adaptive Deformable R-CNN framework for multi-class object detection. Specially, by analyzing the different roles that IoU can play in different parts of the network, we propose an IoU-guided detection framework to reduce the loss of small object information during training. Besides, the IoU-based weighted loss is designed, which can learn the IoU information of positive ROIs to improve the detection accuracy effectively. Finally, the class aspect ratio constrained non-maximum suppression (CARC-NMS) is proposed, which further improves the precision of the results. Extensive experiments validate the effectiveness of our approach and we achieve state-of-the-art detection performance on the DOTA dataset.
引用
收藏
页数:22
相关论文
共 26 条
[1]  
[Anonymous], POSITION DETECTION D
[2]   Towards Multi-class Object Detection in Unconstrained Remote Sensing Imagery [J].
Azimi, Seyed Majid ;
Vig, Eleonora ;
Bahmanyar, Reza ;
Koerner, Marco ;
Reinartz, Peter .
COMPUTER VISION - ACCV 2018, PT III, 2019, 11363 :150-165
[3]   DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs [J].
Chen, Liang-Chieh ;
Papandreou, George ;
Kokkinos, Iasonas ;
Murphy, Kevin ;
Yuille, Alan L. .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2018, 40 (04) :834-848
[4]   Learning Rotation-Invariant Convolutional Neural Networks for Object Detection in VHR Optical Remote Sensing Images [J].
Cheng, Gong ;
Zhou, Peicheng ;
Han, Junwei .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2016, 54 (12) :7405-7415
[5]   Deformable Convolutional Networks [J].
Dai, Jifeng ;
Qi, Haozhi ;
Xiong, Yuwen ;
Li, Yi ;
Zhang, Guodong ;
Hu, Han ;
Wei, Yichen .
2017 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV), 2017, :764-773
[6]  
Dai J, 2016, PROCEEDINGS 2016 IEEE INTERNATIONAL CONFERENCE ON INDUSTRIAL TECHNOLOGY (ICIT), P1796, DOI 10.1109/ICIT.2016.7475036
[7]  
Deng J, 2009, PROC CVPR IEEE, P248, DOI 10.1109/CVPRW.2009.5206848
[8]   Geospatial Object Detection in High Resolution Satellite Images Based on Multi-Scale Convolutional Neural Network [J].
Guo, Wei ;
Yang, Wen ;
Zhang, Haijian ;
Hua, Guang .
REMOTE SENSING, 2018, 10 (01)
[9]   An Efficient and Robust Integrated Geospatial Object Detection Framework for High Spatial Resolution Remote Sensing Imagery [J].
Han, Xiaobing ;
Zhong, Yanfei ;
Zhang, Liangpei .
REMOTE SENSING, 2017, 9 (07)
[10]  
Joseph RK, 2016, CRIT POL ECON S ASIA, P1